Eighty percent of the world's HIV infections are in sub-Saharan Africa. In Lusaka, the capital of Zambia, 85% of pregnant women are married and 47% are in. couples with at least one HIV+ partner (26% concordant positive, 21% discordant). It will be years before short course antivirals are widely implemented and many children who escape infection will be left orphaned. There are 360,000 AIDS orphans in Zambia, a country of 9 million, and 35, 000 HIV+ women deliver each year. Prevention of unplanned pregnancy in couples with HIV can reduce pediatric AIDS, HIV orphans, and the family consequences of parental illness and death. We have previously HIV tested and counseled 10,000 couples from Lusaka for our ongoing NIH funded studies of heterosexual HIV transmission. In a pilot randomized controlled trial of family planning promotion with 251 couples, we observed a marked increase in injectable and oral hormonal contraceptive uptake. The attrition rate was high, however, and compliance with oral contraception was poor. Although incident pregnancy was lower with the intervention, the magnitude of the impact was far less than anticipated. Effective, user-- independent methods such as the IUD and Norplant had lower attrition rates but they are unfamiliar to providers and clients and few women elected to use them. In this study, we will enroll 5000 concordant HIV+ and discordant couples in a randomized factorial trial of two interventions to reduce unplanned pregnancies. The first intervention will promote more effective contraception by placing user independent methods first in the educational message hierarchy and employing positive message framing. The second intervention will help couples plan for the consequences of their illness and death. This will include assisting husband and wife to work together and prepare a will, choose a guardian, and make a financial plan. By focusing on the cost of educating existing children and on the need to plan for their future care, we will encourage couples to reflect on the implications of future childbearing. We hypothesize that this will result in a decision to limit fertility. A standard family planning service will be compared with either or both interventions with respect to impact on incident pregnancy, contraceptive choice and pattern of use, psychosocial and behavioral variables, and future planning actions. Cost- effectiveness will be determined with methods developed jointly by experts in the fields of HIV therapy in Africa and contraception.